A
DIACHRONIC ANALYSIS OF THE PROGRESSIVE ASPECT IN
B
LACK
S
OUTH
A
FRICAN
E
NGLISH
C.M. PIOTROWSKA
Dissertation submitted in fulfilment of the requirements for the Degree Master of Arts in English at the Vaal Triangle Campus of the North-West University.
Supervisor: Prof. Bertus Van Rooy
i
concerning diachronic changes in non-native varieties of English, Schneider’s (2007) indigenous strand. Such a theory must take into account language contact as well as theories pertaining to Second Language Acquisition. This dissertation describes the diachronic changes which occur in one feature of Black South African English (BSAE), namely the progressive aspect. Current synchronic research on BSAE suggests that the progressive aspect is overextended to include stative verbs. This dissertation aims to evaluate the value of this hypothesis, and determine whether there is evidence of this overextension in diachronic data. In order to observe instances of language change, a diachronic corpus of BSAE was complied. This corpus consisted of data from letters, fiction, and newspapers ranging from the 1870s until the 2000s. Using this corpus, analyses were performed in order to determine the frequency of progressive use, the distribution of aktionsart categories, and which temporal meanings were denoted by stative verbs and activity verbs used in the progressive aspect. These analyses were then repeated for data taken from the historical corpus of White South African English (WSAE), in order to ascertain whether changes in BSAE developed parallel to WSAE, or if it follows its own trajectory of change. One further analysis was conducted on BSAE: the aspectual categories of 71 state verbs were analysed in order to determine whether speakers of BSAE indeed do overextend the –ing progressive form onto stative verbs describing states and qualities.
The results indicate that the frequency of verbs used with the progressive aspect increases for BSAE, but decreases for WSAE. The comparison of aktionsart distribution indicates that BSAE has a higher overall frequency of communication verbs and accomplishments, while WSAE has a higher frequency of activity verbs, verbs denoting the future, and stative verbs. Achievements are used as expected by both varieties. The analysis of the temporal meanings of stative verbs indicates that speakers of WSAE denote the prototypical short duration associated with the progressive aspect for 74.83% of the verbs, while BSAE uses an extended temporal meaning for 46.86% of the verbs, indicating that speakers of BSAE more likely to overextend the temporal duration of stative verbs than WSAE speakers. The temporal meanings for activities
ii
duration in both varieties may be due to language change in general, while the additional temporal meanings for statives in BSAE are a result of substrate transfer. Keywords: Black South African English, White South African English, diachronic change, progressive aspect, stative verbs, activity verbs, corpus.
iii
I would first and foremost like to thank my supervisor, Prof Bertus Van Rooy, for his patience, his guidance, his overall brilliance in the field of linguistics, and for letting me hoard several of his books for the duration of this study.
I would like to thank all those who assisted me in my data collection for the BSAE historical corpus. Particularly the kind people at the National English Literary Museum in Grahamstown, the Africana Library and Sol Plaatje Museum in Kimberley, The National Archives in Pretoria, the National Library of South Africa in Cape Town and the Special Collections library at the University of Cape Town.
A word of thanks must also go to the audience members who attended a paper based on this work at the Joint SAALA/SAALT/LSSA conference in June 2014 at the University of Witwatersrand, for their very helpful comments. People who come to mind include Rajend Mesthrie, Lutz Marten, Jochen Zeller, Mark de Vos, and Will Bennett. In addition, I would like to thank Elsabe Taljard, who was kind enough to assist me in glossing examples of Northern Sotho.
Last but not least, I would like to thank my friends and family for their support, and for entertaining my ideas even when they did not know what I was talking about.
I hereby graciously acknowledge the financial assistance of the National Research Foundation (NRF) of South Africa, Grant no. 78795. All opinions expressed within this dissertation are my own and should not be attributed to the NRF.
iv Chapter 1 Introduction ... 1 1.1 GENERAL AIM ... 1 1.2 PROBLEM STATEMENT ... 2 1.3 RESEARCH QUESTIONS ... 2 1.4 OBJECTIVES ... 3 1.5 METHODS ... 4
1.6 OUTLINE OF THE DISSERTATION ... 4
Chapter 2 Theoretical Preliminaries ... 7
2.1 LANGUAGE EVOLUTION AND CHANGE ... 8
2.1.1 Defining language evolution and change ... 8
2.1.2 Innovation ... 9
2.1.2.1 Languages in contact ... 10
2.1.3 World Englishes ... 12
2.1.3.1 Three Circles Model ... 12
2.1.3.2 The Dynamic Model ... 13
2.1.3.3 Situating South African English within the Dynamic Model ... 15
2.2SECOND LANGUAGE ACQUISITION ... 16
2.2.1 What is a Second Language? ... 17
2.2.2 Theories of SLA ... 18
2.2.2.1 Krashen’s Monitor theory ... 18
2.2.2.2 The Identity Hypothesis ... 18
v
2.2.2.5 Pidginization Theory ... 20
2.2.3 The Six Dimensions of SLA ... 21
2.2.3.1 Propensity ... 21
2.2.3.2 Language Faculty ... 22
2.2.3.3 Access ... 22
2.2.3.4 Structure of the Process ... 23
2.2.3.5 Tempo of Acquisition ... 23
2.2.3.6 End-state ... 23
2.2.4 The Four Problems for Learners ... 24
2.2.4.1 The problem of analysis ... 25
2.2.4.2 The problem of synthesis ... 25
2.2.4.3 The embedding problem ... 25
2.2.4.4 The matching problem ... 26
2.2.4.5 Temporality ... 26
2.3 WORLD ENGLISHES AND SECOND LANGUAGE ACQUISITION ... 28
2.3.1 Merging Contact Linguistics with SLA ... 28
2.3.2 Language Transfer ... 30
Chapter 3 Aspect ... 33
3.1 ASPECT IN THE ENGLISH LANGUAGE ... 34
3.1.1 Grammatical aspect: perfectivity and imperfectivity ... 34
3.1.2 Lexical aspect ... 38
3.1.2.1 Aktionsart ... 38
vi
3.1.3.1 Prototypical temporal meaning ... 47
3.1.3.2 Extended temporal meanings ... 48
3.1.3.3 The futurate and susceptibility to change... 52
3.1.3.4 Pluractionality ... 53
3.1.3.5 Towards a synthesis ... 54
3.2 ASPECT IN THE BANTU LANGUAGES ... 57
3.2.1 The Bantu Languages: Zone S ... 57
3.2.2 Traditional Literature on Aspect in the Zone S Languages ... 61
3.2.3 Aspect in South African Languages ... 62
3.2.3.1 Northern Sotho ... 62 3.2.3.2 Southern Sotho ... 64 3.2.3.3 Tswana ... 65 3.2.3.4 Zulu ... 66 3.2.3.5 Northern Ndebele ... 67 3.2.3.6 Tsonga ... 67 3.2.3.7 Venda ... 68
3.2.3.8 The stative verbs in Bantu languages ... 68
3.2.4 The rise of the persistitive ... 69
3.3 THE ASPECT HYPOTHESIS ... 71
3.4 THE PROGRESSIVE ASPECT IN WORLD ENGLISHES ... 73
Chapter 4 Methodology ... 78
4.1 CORPUS-BASED APPROACH ... 80
vii 4.2.2 BSAE in writing ... 81 4.2.3 Sources ... 85 4.3 BSAE CORPUS ... 86 4.4 WSAE CORPUS ... 88 4.5ANALYSIS ... 90 4.5.1 Wordsmith Tools ... 91
4.5.2 Manually identify the progressives to determine frequency... 91
4.5.3 Identifying the Aktionsart ... 92
4.5.4 Temporal meanings of stative and activity verbs ... 95
4.5.5 The progressive aspect in WSAE ... 100
4.5.6 State verbs and their aspectual categories ... 100
4.6 STATISTICAL MEASURES ... 102
4.7 SUMMARY ... 103
Chapter 5 Results ... 105
5.1 BSAEHISTORICAL DATA ... 105
5.1.1 Frequency of verbs used in the progressive aspect ... 106
5.1.2 Aktionsart categories in BSAE ... 108
5.1.3 Distribution of state verbs according to morphology and aspect ... 112
5.1.4 The temporal meanings of stative verbs in BSAE ... 114
5.1.5 The temporal meanings of activity verbs in BSAE ... 118
5.1.6 Summary of BSAE statistically significant findings ... 121
5.2 WSAE HISTORICAL DATA ... 123
viii
5.2.3 The temporal meanings of stative verbs in WSAE ... 128
5.2.4 The temporal meanings of activity verbs ... 130
5.2.5 Summary of WSAE statistically significant findings ... 133
5.3 COMPARISON OF WSAE AND BSAE DATA ... 134
5.3.1 Frequency of the progressive aspect ... 135
5.3.2 Aktionsart distribution across WSAE and BSAE ... 137
5.3.3 The temporal meanings of stative verbs in WSAE and BSAE ... 143
5.3.4 The temporal meanings of activity verbs in WSAE and BSAE ... 147
5.3.5 Summary of statistically significant differences between WSAE and BSAE ... 151
Chapter 6 Conclusion ... 154
6.1 INTERPRETATION OF RESULTS ... 154
6.1.1 Frequency of the progressive ... 154
6.1.2 Aktionsart ... 156
6.1.3 Temporal meanings ... 158
6.2CONCLUSION ... 161
6.2.1 Research Questions and Objectives ... 161
6.2.2 Limitations ... 162
6.2.3 Recommendations for future studies... 163
Appendix 1 ... 164
Appendix 2 ... 166
ix
List of Figures (short list)
Figure 1 Grammatical aspect ... 36
Figure 2 Lexical aspect ... 41
Figure 3 Schematic representation of temporal meanings of the progressive ... 49
Figure 4 Temporal meanings of the progressive when combined with dynamic verbs .. 55
Figure 5 A map of the Niger Congo languages ... 58
Figure 6 A map of Guthrie's Zones ... 59
Figure 7 Grammatical aspect in Northern Sotho ... 64
Figure 8 Grammatical aspect in Zone S Bantu languages ... 70
Figure 9 Flow chart which assists in determining aktionsart categories ... 94
Figure 10 Flow chart which assists in determining temporal meanings of activity and stative verbs ... 98
Figure 11 Frequency of verbs used in the progressive aspect in BSAE ... 106
Figure 12 Frequency of verbs used in the progressive aspect in BSAE Letter data .... 107
Figure 13 Aktionsart distribution in BSAE, normalised per 100,000 words. ... 109
Figure 14 Percentage of temporary, on-going, and unlimited states in BSAE ... 115
Figure 15 Percentage of temporary, regular, on-going, and unlimited activities in BSAE ... 119
Figure 16 Frequency of verbs used in the progressive aspect in WSAE ... 124
Figure 17 Frequency of verbs used in the progressive aspect in WSAE Letter data ... 124
Figure 18 Aktionsart distribution in WSAE, normalised per 100,000 words... 126
Figure 19 Percentage of temporary, on-going, and unlimited states in WSAE ... 129
Figure 20 Percentage of temporary, regular, on-going, and unlimited activities in WSAE ... 131
Figure 21 Comparison of progressive frequency in WSAE and BSAE ... 136
Figure 22 Comparison of the progressive frequency for letter data ... 136
Figure 23 Aktionsart distribution according to percentage for WSAE and BSAE newspaper data ... 138
x
Figure 25 Aktionsart distribution according to percentage for WSAE and BSAE letter
data ... 142
Figure 26 Comparison of temporary, on-going, and unlimited states in WSAE and BSAE ... 146
Figure 27 Comparison of temporary, regular, on-going, and unlimited activities in WSAE and BSAE ... 150
List of Tables (short list)
Table 1 Zone S Bantu languages in South Africa ... 60Table 2 Composition of the BSAE data used for this study ... 88
Table 3 Composition of the WSAE data used for this study ... 90
Table 4 Log-likelihood scores ... 102
Table 5 Raw Aktionsart distribution in BSAE ... 108
Table 6 Aspectual categories of 16 state verbs in BSAE ... 113
Table 7 Temporal meanings of the progressive when combined with stative verbs in BSAE ... 115
Table 8 Temporal meanings of the progressive when combined with activity verbs in BSAE ... 118
Table 9 Raw aktionsart distribution in WSAE ... 125
Table 10 Temporal meanings of the progressive when combined with stative verbs in WSAE ... 128
Table 11 Temporal meanings of the progressive when combined with activity verbs in WSAE ... 131
xi (10) ……….Word Class 10 1………...First Person ADV……….….Adverbial COP………..…Copula IPFV……….Imperfective N……….…..Neutral PL………....Plural PROG……….….Progressive SBJ………..Subject SG………..Singular
1
C
HAPTER1
I
NTRODUCTIONSouth African English comprises five main ethnic Englishes; namely White South African English (henceforth WSAE) which is spoken mainly as a first language, Indian South African English which has developed into a first language variety, and three second language varieties termed Black, Coloured, and Afrikaans South African English (Mesthrie, 2010:5). Each of these varieties are sharply differentiated in both accent and syntax due to the diverse phonological and grammatical features they contain (Mesthrie, 2010:5). In recent years much attention has been given to Black South African English (henceforth BSAE) as a non-native variety of English, with research being conducted on its phonological, grammatical, lexical, and discoursal features (Gough, 1996; De Klerk, 2003, 2006; Mesthrie, 2004; Van Rooy, 2004, Siebers, 2007; Minow, 2010). BSAE can be defined as the variety of English spoken by people who have an indigenous African language as their first or home language (Minow, 2010:6), and typically acquire English as a second language through primary school education.
1.1 GENERAL AIM
The general aim of this dissertation is to investigate how the progressive aspect is used by speakers of BSAE, in comparison to speakers of WSAE, and whether any diachronic developments and changes can be attested in the data. The overarching aim of this dissertation is to describe and explain a feature of a BSAE, as a non-native variety of English, in order to set the stage for future research into the developments of features which occur in other native varieties, so that a theory of diachronic change in non-native varieties may be postulated. A theory of diachronic changes in non-non-native varieties of English, which can outline prototypical change through theories of language contact and second language acquisition, does not yet exist. In achieving this aim, the research presented in this dissertation can serve as a foundation from which such a theory can be postulated.
2
1.2 PROBLEM STATEMENT
The current research on diachronic language change focuses on changes which occur within, and between, native varieties of English. However, no theories documenting diachronic changes exist for non-native varieties of English. One reason for the lack of theoretical information pertaining to diachronic changes in non-native varieties of English is the lack of diachronic corpora, which can be used to investigate the developments of salient features found across non-native varieties of English. Consequently, a call has been made for more descriptive studies profiling diachronic changes (Noël et al., 2014). One feature of non-native varieties of English that has received much attention in research on BSAE is the so-called overextension of the progressive aspect to include stative verbs. Current studies on the progressive are all synchronic, and focus mainly on the description of stative verbs when used with the progressive aspect. Therefore, diachronic studies which examine this feature’s saliency throughout history are needed. Furthermore, research is needed on how dynamic verbs are used with the progressive aspect as well, as these have not yet been profiled in much detail. A further issue is that the progressive aspect in BSAE has not yet been compared to the use of the progressive in WSAE, particularly diachronically.
1.3 RESEARCH QUESTIONS
Based on the gaps identified in the problem statement above, this dissertation has three research questions:
1) Have there been any changes in the frequency and meaning of the progressive aspect in BSAE since the colonial period (1880s) until contemporary times (2000s)?
2) How do the findings from the BSAE (Schneider’s (2007) indigenous strand) data compare to those of WSAE (Schneider’s (2007) settler strand)?
3) Which theoretical factors pertaining to language change are required to explain the changes found in the empirical data?
3
1.4 OBJECTIVES
The primary objective of this dissertation is to provide an account of diachronic changes of the progressive aspect in BSAE. The majority of current studies examine the specific use of statives as progressives in contemporary data, while only a few (Siebers, 2007; Minow, 2010) deal with the progressive aspect as a whole. However, we do not know how long this feature has been present in BSAE nor how the progressive aspect has developed over time, making it essential to do research on diachronic changes in BSAE. This dissertation will attempt to rectify this by analysing the frequency of verbs used with the progressive aspect in BSAE, as well as the aktionsart distribution of these verbs in order to determine whether there have been changes in frequency and meaning of the progressive aspect. Furthermore, this dissertation will analyse the temporal meanings of the stative verbs, as well as dynamic activity verbs, in order to ascertain whether speakers of BSAE use a longer duration than is prototypical for the progressive aspect.
It is important to know not only how the progressive aspect has developed, but also whether it has developed in accordance with other varieties of South African English. The second objective is to relate and contextualise the findings in the light of other South African varieties, particularly by observing whether developments in Schneider’s (2007) IDG strand (BSAE) correspond to the STL strand (WSAE). Thus, the focus is on whether BSAE and WSAE develop along a parallel path, or if each variety moves along its own trajectory.
The third objective of this study is to account for the changes (or the lack thereof) which are found while examining the data, using existing knowledge on theories of language change and evolution, as well as second language acquisition. Specifically, the importance of language contact and language isolation to the development of new features is stressed, as well as theories pertaining to language transfer from contact linguistics (transmission and diffusion) and SLA (interlanguage and substrate influence).
4 1.5 METHODS
In order to investigate how language changes diachronically, this dissertation makes use of a corpus-based approach. Corpus linguistics has provided researchers with an effective way of analysing patterns of use, ensuring that frequencies of features are reported accurately, and confirming whether a feature is truly characteristic of a variety (Biber et al., 1998:4; Minow, 2010:1). A corpus-based approach makes a quantitative study of this size feasible, as a wide variety of texts can be searched and examined for specific features in a fraction of the time that it would take to go through each text manually.
1.6OUTLINE OF THE DISSERTATION
This dissertation begins, in Chapter 2, with an overview of the existing body of literature surrounding theories of language contact and change, as well as Second Language Acquisition (SLA). This chapter is focused on describing language evolution and change by means of discussing how language contact and language isolation drive linguistic innovation. Thereafter, the discussion turns to review processes of selection and new dialect formation. World Englishes are subsequently reviewed by means of exploring Kachru’s Three Circles Model and Schneider’s Dynamic Model, in order to illustrate that BSAE represents the indigenous (IDG) strand, while WSAE can be said to represent the settler (STL) strand. Subsequently, SLA is reviewed with a focus on the processes of acquiring a second language, and the tasks that learners need to overcome to reach native-like competence. Finally, this chapter draws together the two lines of thought and illustrates how theories of language contact and SLA can be used together in order to describe non-native varieties of English. This chapter is important to the dissertation as it explains the theoretical constructs which underlie my own thinking, and also highlights existing gaps in the literature regarding the diachronic development of non-native varieties.
Chapter 3 of this dissertation describes grammatical and lexical aspect, or aktionsart, of Standard English and Zone S Bantu languages. This chapter therefore describes the feature which is investigated in this dissertation, namely the progressive aspect, and contributes to the classification of the categories for my analysis. In the first section, the
5
perfective versus imperfective distinction is discussed in relation to Standard English, followed by a review of the aktionsart categories used to describe the function of verbs. Thereafter, the inherent temporal meanings which are denoted by verbs used in the progressive aspect are discussed. The second section focuses on the aspectual categories found in Zone S Bantu languages, and highlights the additional imperfective category present in Bantu aspect, namely the persistitive aspect. The third section reviews current literature on the changes in the use of the progressive aspect for non-native varieties of English, with a particular focus on the literature surrounding BSAE. The gaps in the current description of the progressive aspect in BSAE are subsequently highlighted.
Chapter 4 of this dissertation describes my methodology. The chapter first explores the advantages of using a corpus-based approach to study diachronic changes in language. Thereafter, this chapter addresses issues in compiling a historical corpus of a non-native variety, namely data collection, corpus design, and possible sources of data. Next, the corpora used in this dissertation, the diachronic corpus of BSAE and the diachronic corpus of WSAE, are described. Subsequently, the discussion turns to outline the types of analyses that were performed on the BSAE and WSAE corpora; namely the 1) determination of the frequency of the verbs used with the progressive aspect, 2) the distribution of the aktionsart categories for the verbs with the progressive aspect, 3) the identification of the temporal meanings for stative verbs, and 4) the identification of the temporal meanings for the activity verbs. Since the feature analysed here is typically defined as the overextension of the progressive aspect to include state verbs, one extra test was performed on BSAE alone, namely the classification of aspectual categories used in combination with state1 verbs, in order to determine whether the claim of overextension is accurate. Finally, the statistical measure used to calculate the significance of any differences found in the data is outlined.
Chapter 5 of this dissertation presents the findings from the analyses described in Chapter 4, in three sections. The first section presents the results from the analysis of
1 State verbs are considered a subcategory of stative verbs, and are used to describe private states,
intellectual states, states of emotion and perception, and states of bodily sensation (Quirk et al., 1985:203).
6
the historical BSAE corpus, in the order in which the analyses where performed. The second section presents the results from the analysis of the historical WSAE corpus. Finally, the last section aims to compare the results from the WSAE to those of the BSAE corpus, in order to identify any differences and similarities in the frequency of the progressive aspect across each time period, and for each of the newspaper, letter, and fiction subcorpora. Thereafter, the aktionsart distribution is compared in order to determine whether or not BSAE overextends the progressive aspect to stative verbs, or whether BSAE distributes the aktionsart in the same way as WSAE. In addition, the temporal meanings of the stative and activity verbs for WSAE and BSAE are compared, in order to determine whether speakers of BSAE denote a longer duration than what is prototypical for the progressive aspect in Standard English.
Chapter 6 of this dissertation critically discusses the findings from the analysis of BSAE and WSAE, with particular focus on the results which arise from the comparison of the two varieties. These findings are discussed in light of the theoretical constructs explored in Chapter 2 of this dissertation, namely language contact and change and SLA. The conclusion of this chapter serves as a conclusion to the dissertation. The conclusion returns to the three research question posed in this introductory chapter, along with the objectives of the study in order to determine whether this dissertation achieved all that it set out to do. Finally, the limitations or this dissertation are reviewed, and recommendations for future studies are made.
7
C
HAPTER2
THEORETICAL PRELIMINARIESThis chapter presents an overview of the existing body of literature surrounding the theoretical and contextual concerns relevant to the study of the progressive aspect in BSAE on a diachronic level. This chapter critically reviews the theories and frameworks which are adopted in this dissertation, and is divided into three sections.
This chapter begins with a discussion of language evolution and change, in order to explain the existence of the different varieties of English. Specifically, the factors which influence language evolution and change are first discussed, followed by an explication of the effects that language contact and isolation have on innovation. Thereafter, the discussion turns to examine theories of World Englishes. The Three Circles model proposed by Braj Kachru is first examined, followed by an exploration of The Dynamic Model of the Evolution of Post-Colonial Englishes (henceforth the Dynamic Model) by Edgar Schneider. This section ends by briefly situating South Africa within the Dynamic Model.
Subsequently, this chapter examines first language acquisition and second language acquisition (henceforth SLA), as the type of language acquisition plays a role on an individual speaker’s competence and proficiency in a particular language. Since English is learnt as a second language by black children in South Africa, the notion of second language acquisition makes up the crux of this section. This section explores theories concerning SLA, six dimensions of language acquisition which determine and characterise the process of acquisition, four problems that SLA learners frequently encounter, and the major differences between first language acquisition and SLA. Within this discussion, attention will be drawn to the issue of temporality, a part of the embedding problem that many learners are faced with, as it is here where the issue of aspect lies.
The final section of this chapter draws together language contact and SLA, showing how SLA can be applied to research on language contact and change, particularly when dealing with World Englishes. The errors versus features debate is reviewed in order to
8
highlight the issues of describing features of non-native varieties in deficiency terms. Finally, the issue of language transfer is discussed due to its particular importance for this dissertation. Attention is drawn to the difference in description of transfer for contact linguistics and SLA, the former focusing on transfer between individuals in groups, the latter how individuals transfer features from their native language into a target language.
2.1LANGUAGE EVOLUTION AND CHANGE
This section introduces theories of language change and languages contact, which as the primary theories in which this dissertation is grounded, serve as a point of departure for my own thinking. The discussion is divided into three parts, and begins with a brief definition of the concept of language evolution in Section 2.1.1, which is used in the literature to refer to changes, and explores the internal and external factors which influence change. Thereafter, the notion of innovation is explored in the Section 2.1.2, by means of discussing languages in isolation and languages in contact, and by surveying the processes of new dialect formation as well as competition and selection. This discussion is followed by an explication of the processes of language transfer, reviewing the role of adults and children in feature transmission, and briefly discussing substrate influence. Section 2.1.3 deals directly with theories pertaining to World Englishes, namely Kachru’s Three Circles model and Schneider’s Dynamic Model, and situates South African English within these models.
2.1.1 Defining language evolution and change
Linguists work on the premise that languages are dynamic, constantly undergoing change and evolving, so long as there are still speakers of the language (Mufwene, 2013:206). Language evolution is a slow process that takes place across many generations which consists of two stages: mutation, or innovation, and selection. In their book Evolutionary Linguistics, McMahon and McMahon (2012) deal with the genetic evolution of language, and note that when linguists use the term “evolution” they mean something a little different, and examine 4 cases: 1) Samuels’ (1972) use of evolution evoke a framework for understanding language change, and consists of a spoken chain and system which are connected by the process of selection. 2) Mufwene (2001) uses
9
concepts from ecology and population genetics in an approach to creolisation and language change. 3) Croft (2000; 2006) draws on David Hull’s (1988; 2001) generalised analysis of selection and replication (imperfect copying of a feature). Croft applies this to language change, noting that speakers will produce different variants of “linguemes” (structures in utterances) in different contexts, the frequency of which will alter over time, thereby causing change. 4) Blevin’s (2004) book Evolutionary Phonology deals with the development of sound patterns across time, and that patterns may recur in languages which belong to different families. Thus, when linguists speak of language evolution they essentially mean language change.
Language change is a complex process driven by linguistic (internal) and social (external) factors (Trudgill, 2004; Schneider, 2007; Mufwene, 2008; Labov, 2010). Internal changes are driven by linguistic processes such as grammaticalization. External factors include age, gender, race and ethnicity, and social class (Stockwell, 2007; Labov, 2010). The latter two factors are particularly important in the context of South Africa (Lanham & MacDonald, 1979; Coetzee-van Rooy & Van Rooy, 2005; Bekker & Eley, 2007; Mesthrie, 2010), as the varieties of South African English are shaped by the different racial groups within South Africa. Language change is generally studied diachronically, through the use of corpora that can stretch back centuries, but can also be studied synchronically by profiling the speech patterns of people in various age groups, representing different generations of speakers (Stockwell, 2007:15). Language change is evidenced by innovations in the language, on a broader level wherein borrowing and koinéisation occurs, and on a more specific level wherein existing features develop new meanings and new uses.
2.1.2 Innovation
Linguistic innovation can occur in phonology, morphology, syntax, grammar, and in meaning. Innovation can occur both in situations of isolation and language contact (Labov 2010). Languages left in isolation will develop in their own direction, and typically become increasingly complex. With no external interference, competition between features is not as fierce, which results in the accumulation of irregularity. This is evidenced by the fact that the most morphologically complex languages have been
10
historically the most isolated, such as Icelandic and Faroese. In contrast, languages that are in contact with one another may share features, but the new feature will frequently replace an older one, which results in simplification. The process of language change is, of course, far more complex with a multitude of processes, some of which are now discussed in connection to language contact.
2.1.2.1 Languages in contact
Competition and selection are integral parts of language change (Mufwene, 2008:132). In the event that two linguistic communities converge, they will interact and share linguistic features; these features will then compete. Mufwene (2008:117) explains that according to the feature-pool approach, in times of language contact members of a community will interact and contribute different phonological, lexical, or grammatical features to the feature pool. These features are the result of mutations or deviations from the norm, and can be classified as innovations. The community will then unconsciously2 select features from the feature pool and incorporate these in their speech. However, the same feature is not necessarily chosen by each individual, or the feature could be used in different ways. Selection is possible because language, unlike genetics, is acquired piecemeal (Mufwene, 2008:120) and because individual speakers will make accommodations for other speakers. Linguistic conventions therefore arise from - and are enforced through - cultural interaction (Tomasello, 2008:296), they are based in a shared common ground and can be motivated by a need for communication, collaboration and identity alignment. Factors that play a role in selection include frequency and salience of a feature, markedness, historical depth, transparency and regularisation, similarity of features, the status of the speakers, and the identity with which speaker wish to align (Schneider, 2007:110-111). Labov (2010:244) notes that there is a possibility that ideology is a driving force behind change, but also a barrier that inhibits its expansion, since people will not speak in the same way unless they share a sense of community. Thus people with the same ideologies will identify with one another and speak the same dialect, but separate themselves linguistically from people who do not share their views.
2 While the selection process is largely unconscious, sometimes speakers may consciously select a
11
Language contact leads to the formation of new dialects (Trudgill, 2004:84-89) Trudgill (2004:84-89) identifies six mechanisms which are involved in the formation of new dialects. The first process is called mixing, which occurs when people of different dialects of mutually intelligible languages come together in one location. The second process is called levelling, and involves the loss of minority features, and is determined by which group of people will accommodate other speakers. The third process is termed unmarking, a type of levelling which involves the degree of linguistic markedness ad regularisation, and inevitably dictates which features may be lost. The fourth process is interdialect development, where newly created features result from the interaction of differing dialects. The fifth process is reallocation, wherein any surviving features from the original dialects are reassigned as specific social class, stylistic, or allophonic variants. Taken together, these five processes can be called koinéisation (Trudgill, 2004:89). For new dialect formation to be successful, a sixth process is necessary. This final process, focusing, allows the new dialect to acquire norms and stability.
One important factor leading to the differentiation of dialects and languages is isolation. As noted above, communities will unconsciously create and select the features to incorporate into their speech; but these features will differ for different communities, isolation thereby allows each community to develop along its own path. Regardless of their separation, some languages will develop along similar patterns, a phenomenon known as “drift” (Trudgill, 2004:131). Languages that are geographically separated but which derive from the same source will have their own trajectory, but will most likely pass through the same phases since some of these changes were already in place when the languages separated (Trudgill, 2004:132).
At times, the normal progression of a language can be interrupted for about one generation, after which the development of linguistic changes may continue. This phenomenon is generally known as colonial lag (Trudgill, 2004:34), wherein a delay in the normal progression of linguistic change can be observed in new colonial situations, with the consequence that there is no peer-group dialect for the children to acquire. As a result, children’s speech is more influenced by their parents’ than it would be otherwise.
12 2.1.3 World Englishes
The theory of new dialect formation cannot account for what happens when languages which are not mutually intelligible come into contact. This section focuses on language contact resulting from British colonisation, and the varieties of English which emerged as a consequence of colonisation. First, the different types of Englishes are reviewed with Kachru’s Three Circles model in mind. Thereafter, Schneider’s Dynamic Model is briefly explained.
2.1.3.1 Three Circles Model
The notion that there is a difference between native and non-native varieties of English first emerged in the distinction of Englishes into three categories depending on the functional and political role of English in a given country (Schneider, 2003:236). The first differentiation was between countries in which the people spoke English as a Native Language (ENL), English as a Second Language (ESL), or English as a Foreign Language (EFL). In 1985, Braj Kachru created the Three Circles model, which consisted of three varieties of English that corresponded to the ENL/ESL/EFL distinction (Schneider, 2003:237; Van Rooy & Terblanche, 2010:358). Kachru differentiated between Inner Circle varieties (ENL) such as British English and American English, Outer Circle (ESL) countries wherein English is used in public life for communication among non-native speakers, such as India or South Africa, and Expanding Circle (EFL) countries wherein English is mainly used for contact with foreigners but hardly in the country itself, such as Japan.
Since then, research on non-native varieties of English expanded, along with the terminology used to describe these varieties3. To clarify my own usage: the term ‘native variety of English’ refers to varieties of the Inner Circle, ENL, or Standard English4; whereas the term ‘non-native variety of English’ refers to varieties of the Outer Circle or
3 In addition to the ones listed here, linguists also distinguish: Metropolitan Standard Englishes, Colonial Standard Englishes, Regional/ Social dialects, Pidgin Englishes, Creole Englishes, Immigrant Englishes, Language-shift Englishes, Jargon Englishes, Hybrid Englishes, English as a Lingua Franca, English as an International Language, and English as an Additional Language (Mesthrie & Bhatt, 2008:3-11).
4 Standard Englishes, particularly those of Britain and America, may also contain non-standard dialects. However, these are not essential to my discussion.
13
ESL, as well as the varieties of the Expanding Circle or EFL, and can also be referred to as New Englishes or World Englishes5.
2.1.3.2 The Dynamic Model
The Dynamic Model was proposed by Edgar Schneider in 2003, and later expanded in his 2007 book Postcolonial Englishes: Varieties around the World. The Dynamic Model was informed by theories of language contact, including the impact of multilingualism, SLA, and transfer, as well as theories pertaining to sociolinguistics, social identity, and language evolution (Schneider, 2014:11). The Dynamic Model outlines language development and change in countries that have been colonised, with a focus on the language of both the settlers (henceforth the STL strand) and the indigenous people of the country (henceforth the IDG strand).
According to his Dynamic Model (Schneider, 2003; 2007) which shows the evolutionary stages of New Englishes, English settlers would arrive in a country wherein the indigenous people had no previous exposure to English and only spoke their native language(s). As the settlers come into contact with the indigenous people in the foundation stage, there is koinéization and toponymic borrowing by the settlers, followed by lexical borrowing of terms for fauna and flora and cultural items in stage two (Schneider, 2007). This borrowing occurs when the settlers have more contact with the locals and begin to learn about their customs. Thereafter, there is a third stage of heavy lexical borrowing, until the new variety reaches stabilization in stage four. In stage five, we see the beginning of new dialect birth, with varieties emerging that are group specific (Schneider, 2007).
Schneider (2007) notes that there is a social pre-condition for any linguistic change to happen, and identifies identity construction as an important factor for converging language change. This is clearly seen in the Dynamic Model, in which each stage of development relies on reconstruction of identity. During the foundation stage, the settlers identify themselves with their original nation, while the indigenous community
5 Mesthrie and Bhatt (2008:3) prefer the term English Language Complex (ELC), arguing that New Englishes is sometimes applied to varieties that are not at all new, such as Indian English which is in fact an older variety than Australian English, while the term World English is over-general since it does not encompass Standard varieties of English.
14
retain their local identity. In stage two, the exonormative stabilization, settlers view themselves as an outpost of their original nation, meaning that they acknowledge themselves as settlers, but still identify with England. While in stage three, nativization, members of the STL strand become permanent residents in the country, yet still acknowledging their English origins, in the same way that the IDG strand differentiates themselves as permanent residents of indigenous origin. Thus, in the first three developmental stages, there is a clear separation between the two groups in terms of their identity. It is only in stage four, endonormative stabilization, when the two groups align themselves in terms of identity, and view themselves as members of a new nation. For this alignment to take place, Schneider (2007) suggests that a historical or political event (Event X) must occur to create a sense of community, and feeling of post-independence or self-dependence. As a result of this alignment of identity there is an acceptance of the local norm, and people have a positive attitude towards this, and we also find stabilization of the new variety of English, homogeneity and codification (Schneider, 2007). Finally in stage five, differentiation, identity becomes group specific but remains a part of an overarching new national identity, which results in dialect birth. Therefore, in the last two developmental stages, there is no longer a differentiation between the STL strand and IDG strand but instead an idea of a unified national identity. Once this national identity is in place, people may group themselves in terms of their ideologies and these groups may begin to develop unique features, forming new dialects.
Traditionally, Schneider (2003, 2007) uses the term Post-Colonial Englishes (henceforth PCE) to refer to varieties of Kachru’s Inner and Outer Circle, as this was the focus in his Dynamic Model. However, Schneider’s (2014) more recent work explores the possibility of applying the Dynamic model to Expanding Circle varieties as well. Schneider (2007) provides a few possible causes for why PCEs differ from Standard English: 1) language is closely related to culture, 2) can be explained to an extent by universal laws of ontogenetic second language acquisition and phylogenetic language shift, 3) selections conform to an overarching language type, with innovations being typologically similar. While PCEs may differ from Standard English, there are a striking amount of similarities
15
between the PCEs themselves, regardless of whether the variety arose from a different country and different context (Schneider, 2007:83).
One issue with the application of the Dynamic Model is that most current research focuses on the developments and changes in the STL strand in countries that were previously colonised. There is currently no existing theory of diachronic change for the non-native varieties of English, Schneider’s IDG strand. Change within newly emerging varieties is discussed as emergence of new features via innovation and effects of language contact, but the idea of ongoing internal change within such varieties in not directly addressed in the Dynamic Model, although the possibility is not ruled out (Schneider, 2013 personal communication). Consequently, we do not know what is considered to be prototypical change in a non-native variety, or if changes occur at all. The fact that English is learnt as a second language for speakers of BSAE creates a complication. In the IDG strand it is not predominantly child first language acquisition that shapes features, but adult SLA under contact conditions (Schneider, 2013 personal communication; Schneider, 2013:143).
2.1.3.3 Situating South African English within the Dynamic Model
In applying the Dynamic Model to South Africa, Schneider (2007:188) argues that South African English is entering phase 4, endonormative stabilization. Van Rooy and Terblanche (2010:407) investigate Schneider’s (2007) claim regarding South African English and note that while lexical innovation is fully incorporated and accepted, morphological complex forms are largely but not wholly accepted by speakers of South African English. Thus, there is evidence that South African English is moving beyond phase 3 and into phase 4 (endonormative stabilization) (Van Rooy & Terblanche, 2010:407). Bekker (2009:86) emphasizes that the sub-varieties of South African English should be allowed to develop along their own paths, and places BSAE in phase 3 (nativisation) while in WSAE he observes original settler-based regionalisms instead of phase 5 (differentiation) developments. Following this view, we can identify WSAE as the representation of the language of the STL strand of the Dynamic Model, while BSAE represents the IDG strand.
16
In summary, language change can occur in situations of language contact and isolation. When speakers with different linguistic repertoires interact, they may share features. These features form a feature pool, in which variations must compete and become selected by speakers. If speakers of different dialects converge in one location, a complex process of koinéisation may begin, resulting in the formation of new dialects. The Dynamic Model explores the development of English in a colonial situation, wherein speakers of a Standard variety of English settle in a country where no English exists. The indigenous people of the colonised country come to acquire English through prolonged contact with the settlers, resulting in two distinct strands of English, namely the STL strand, a native-variety of English, and the IDG strand, a non-native variety of English. Current research focuses its attention on how native varieties of English changed and developed diachronically, but little attention is given to the historical development of the non-native varieties of English. One primary difference between the two strands is that in the STL strand it is child first language acquisition that shapes the features, while in the IDG strand it is adult SLA under contact conditions. Therefore it is empirical to consider theories of SLA alongside language contact when attempting to describe developments or changes over time. In the next section, the processes driving SLA will be discussed.
2.2SECOND LANGUAGE ACQUISITION
This section deals with theories pertaining to the process of acquiring a second language, and begins with a brief definition of what is meant by first and second language. Researchers have long held the assumption that there are differences between first and second language acquisition (SLA), however this notion only became systematically addressed in the late 1960s (Meisel, 2011:2-3) In section 2.2.2, five prominent theories of SLA are evaluated; namely Krashen’s monitor theory, identity hypothesis, contrastive hypothesis, theories of learner varieties, and pidginization theory.
A second language can be either tutored through systematic guided language learning, or learnt spontaneously through everyday communication. According to Klein (1990:35-52), the process of acquiring a new language is determined by three factors –
17
propensity, language faculty, and access – and characterised further by three categories – structure, tempo, and end state – which together create the six dimensions of SLA. These dimensions are subsequently discussed in section 2.2.3. While acquiring a new language learners are faced with four tasks which they need to carry out. The topic of section 2.2.4 is the problems that may arise from these four tasks, namely the problems of analysis, synthesis, embedding, and matching. Particular attention is given to the notion of temporality, a part of the embedding problem, which deals with time reference, aspect, and action type.
2.2.1 What is a second language?
All people use language, whether spoken or signed, in order to express themselves and communicate with other members of society. From birth (and perhaps already in the womb) they are exposed to at least one language which they inevitably learn to speak. Later in life, during childhood or adulthood, the individual may be exposed to other languages and decide to learn one or more. At other times, languages are thrust upon them during education. Research has shown that there are differences in the acquisition process depending on whether an individual learns a first, or a second language. To clarify, a language that is taught from birth as a mother tongue is referred to as a first language; even if the individual learns two or more languages at this time (Klein, 1990:3). Alternatively, any language that is learnt later in life can be classified as a second language (Klein, 1990:3),6 whether learned inside or outside a classroom (Ellis, 1997:3). These differences in the acquisition process may account for the different levels of proficiency in a first and second language. Other factors, as can be seen in sections 2.2.3 – 2.2.4, also influence a speaker’s competence in a second language. Corder (1967; in Meisel, 2011:8) lists a few differences between first language acquisition and SLA, most notably that 1) children are more successful in acquiring their first language rather than a second language, 2) first language develops as part of a child’s maturation process, 3) at the onset of SLA, there is already another language
6 This distinction is somewhat hazy, as if is possible for children to begin learning a second language before they have mastered the first. Thus, an exact definition for first language acquisition and SLA is difficult to formulate.
18
present, and therefore 4) the motivation for language learning is different for the first and the second language.
2.2.2 Theories of SLA
There are two major goals of SLA research, namely to describe the process of acquiring a second language, and then to explain this process by means of identifying the external and internal factors which contribute to this acquisition (Ellis, 1997:4). It is clear from the above section that the process of acquiring a language is vastly complex, and differences in this process arise whether the language a person learns is their first (native) language, or a second language acquired later in life. Although these differences have always been noted, serious empirical research to back these findings only began in the 1960s (Meisel, 2011:2-3). This research lead to a number of theories and hypotheses regarding SLA, five of which will be discussed in turn below.
2.2.2.1 Krashen’s monitor theory
Krashen’s monitor theory differentiates between subconscious acquisition and the conscious learning of a language (Klein, 1990:28). Although it is not a model for language acquisition in general, it highlights the way in which the process of acquisition is influenced by conscious awareness. Learning is monitored by the speaker, who consciously corrects and controls their language output and follows grammatical rules (Klein, 1990:28). Alternatively, speakers who acquire language subconsciously are concerned with the content of their utterances (not the form), and aim to communicate meaningfully with other speakers (Klein, 1990:28).
2.2.2.2 The Identity hypothesis
The identity hypothesis proclaims that all types of language acquisition essentially follow the same process with the same laws, regardless of whether a language has been learned before (Klein, 1990:23). This view has not since been widely upheld, since differences between first language acquisition and SLA do exist, and each method of acquisition is in itself highly variable depending on whether the process is tutored or untutored.
19 2.2.2.3 The Contrastive hypothesis
Contrastive hypothesis, on the other hand, asserts that the acquisition of a second language is largely influenced by the structure of an earlier acquired language (Klein, 1990:25). Contrastive Analysis (CA) stemmed from a line of thought expressed by Charles C. Fries in 1945, and was refined by a former student of his, Robert Lado, who made a rather generalised claim that learners tend to transfer forms and meanings from their native language into a new language that they are learning (Meisel, 2011:3-4). Structures which corresponded in both languages are learnt easily with the help of “positive transfer” (Klein, 1990:25; Ellis, 1997:51), while contrasting structures often proved problematic due to the effect of “negative transfer” or “interference”7 (Klein, 1990:25; Ellis, 1997:51). The role of this transfer was, however, overemphasised due to CA’s focus on structures. Acquisition is concerned with the actual comprehension of a language, rather than the replication of a structure or abstract linguistic form (Klein, 1990:25-26). This is not to say that transfer plays no role in SLA, as many learners rely on knowledge of their first language to assist them in comprehending and producing utterances, but this is only one of many resources that are available to them (Klein, 1990:27). Expending on this notion, researchers such as Corder (1967) suggest that the focus of acquisition studies should be not on the production of structures, but instead on the learners.
2.2.2.4 Theories of learner varieties
Learner varieties of a target language represent the learner’s current linguistic repertoire, and several theories have been formulated around this concept (Klein, 1990:29-30). Theories of learner varieties are premised on the notion that 1) all language varieties have an intrinsic systematicity, and 2) the process of language acquisition can be viewed as a series of transitions from one language variety to the next, which in turn reveals an inherent systematicity (Klein, 1990:29). An initial proposal of this kind was put forth by Corder (1967), who noted that in the process of SLA learners make both random mistakes and systematic errors (Meisel, 1990:8), the latter
7 The issue of transfer (interlanguage) is discussed in some detail in Section 2.3.2, taking into account SLA as well as issues pertaining to language contact.
20
revealing the learners’ “transitional competence” (Corder, 1967:166). Arguably the most influential concept developed as a learner variety was the notion of interlanguage, made popular by Selinker in 1972 (Klein, 1990:29; Meisel, 2011:9; Schneider, 2013:143). Interlanguage is determined by the combined factors of type of transfer and strategy of acquisition (Klein, 1990:29). Interlanguage is usually viewed negatively; Thomason (2001:146; in Schneider 2013:144) offers an alternative position in hypothesising a gap-filling approach, in which the learner uses their native language to assist them in gap-filling gaps in their knowledge of the target language.
2.2.2.5 Pidginization theory
Pidginization accounts for the acquisition of a language under conditions of restricted input to the target language. A pidgin can be defined as “a second language which comes into being whenever speakers of a politically, socially, or culturally subordinate language try to acquire some knowledge of a dominant language for specific purposes.” (Klein, 1990:30.), Pidgins can be considered as special cases of spontaneous SLA, and are similar to learner varieties in their structure and emergence. For further discussion of the similarities, refer to Klein (1990:30-32).
In summary, a variety of theories have been proposed with regards to SLA. Krashen’s monitor theory demonstrates that language learning differs if acquisition is tutored or untutored. While identity hypothesis proposes that knowledge of another language is irrelevant to SLA, it is challenged by the contrastive hypothesis which states that learners transfer structures from their native language into the language they are learning. When this transfer is negative, it may give rise to errors in use as a result of interference. Theories on learner varieties focus on learners’ competence during the transition between beginning learning and become a fluent speaker. The process of language acquisition is thus a complex one, determined by six factors which will be subsequently discussed. In cases where a learner’s contact with the target language is severely restricted, speakers of a subordinate language may develop a pidgin in order to communicate with speakers of the dominant language.
21 2.2.3 The six dimensions of SLA
This section arranges the multitude of factors influencing the process of language acquisition along six dimensions, as described by Klein (1990:35-52). These dimensions are comprised of three components which determine the process, and three categories that characterise the process of SLA. Regardless of whether the learner obtains the second language through systematic learning or spontaneous acquisition, the following six dimensions influence their ability to learn the new language with a degree of proficiency.
2.2.3.1 Propensity
The term propensity covers the factors that encourage learners to apply their language faculty to acquiring a new language, and it is the outcome of this interaction that determines their propensity (Klein, 1990:35). This includes four types of motivations. Firstly, the learner may be motivated by social integration. Although this is more important for child first language acquisition, it may also influence adult SLA to a degree as a learner may be motivated to learn the second language in order to align their identity and to successfully integrate themselves into the community of speakers (Norton Pierce, 1995:17; Ellis, 1997:75). This may also have an adverse effect if the learner resists in order to keep the identity associated with their native language (Klein, 1990:36). Secondly, the learner can be motivated instrumentally, and will learn a language in order to satisfy their communicative needs (Klein, 1990:36-37; Ellis, 1997:75) and also for utilitarian purposes (Norton Pierce, 1995:17). The third type of motivation is resultative, since the result of their learning may dictate whether learners become more, or less, motivated to learn after experiencing success (Ellis, 1997:75). The final type of motivation is intrinsic, which requires learners to keep themselves interested and maintain their curiosity in learning a new language (Ellis, 1997:76).
In addition, the learner’s attitude may also influence the process of acquisition, as some learners are more willing than others to communicate with speakers of the target language regardless of their imperfect command over the language and the embarrassment that may come from the experience (Klein, 1990:37-8). The subjective attitude may also influence, consciously or unconsciously, the speaker’s need to learn a
22
new language (Klein, 1990:38). Speakers who are willing to communicate in the new language expose themselves to further contact its speakers, thus enabling them to progress. The final factor of propensity is education. Research has found that even if children are taught the target language at school, a child might not be successful without the first three propensity factors (Klein, 1990:38-39). Education may also provide motivation for learning a language, as the child may learn well in order to achieve social advancement or to avoid punishment from teachers (Klein, 1990:39).
2.2.3.2 Language faculty
Humans have the ability to understand and process languages, and the language processor in our brains can adjust according to the particular linguistic material of a language while we learn (Klein, 1990:39). The functioning of the language processor is dependent on certain biological determinants and the knowledge available to the learner. Biological determinants include our peripheral organs and central nervous system, which enables us to hear speech sounds and decode them, and also to produce utterances (Pinker 1994: 158-170). The knowledge available to the learner can be conscious (contextual), or tacit (Klein, 1990:40-43); and in order to communicate effectively the learner is required to combine contextual and linguistically conveyed information.
2.2.3.3 Access
In order for the language processor to operate, the speaker must first have sufficient access to the target language. The term access covers two components, namely input and opportunities to communicate (Klein, 1990:43-47). The learner receives input from the speech of others and the corresponding contextual information. Although the speech is usually authentic, it can be adjusted by the speaker in attempts to accommodate the language learner (Klein, 1990:45). This may include foreigner talk, simplified registers, and motherese. Although the speaker assumes that this assists in language development, researchers discredit this assumption (Pinker, 1994:39-40). Learners also gain access through opportunities to communicate with native speakers (Klein, 1990:46-47), wherein they can hear the language and have a chance to speak.
23
Propensity, language faculty, and access determine the process of language acquisition. This process can be further characterised by three categories, namely the structure of the process, the tempo, and the end-state. The propensity, language faculty, and access to the target language directly coincide with, and influence, the remaining three factors.
2.2.3.4 Structure of the process
Klein (1990:47) notes that there are two key notions regarding the structure of the process of language acquisition. The first deals with the way in which the various elements that make up proficiency in a language are synchronised. The speaker is expected to make proper use of information from phonological, morphological, syntactic, and lexical domains (Klein, 1990:48) in synchrony, and not learn each in isolation. The second notion deals with the issue of variability, as the structure of acquisition varies across individual learners (Klein, 1990:49). Consequently, SLA cannot be viewed as a uniform process.
2.2.3.5 Tempo of acquisition
Propensity, the language faculty, and access to the target language influence the tempo at which a learner acquires a language (Klein, 1990:50). The issues of propensity and access have a significant impact here, since they can contribute to the acceleration of the learners’ progress (Klein, 1990:50) under the condition that the learner has sufficient access to the language and is prepared to communicate with other speakers. Thus, opportunity and motivation may have a positive outcome in that they accelerate the process. On the other hand, if a speaker’s attitude does not reflect willingness to communicate or the speaker is simply not afforded the chance, it may result in a deceleration of progress. With regards to the language faculty (Klein, 1990:50), an impairment of a biological determinant may also slow down the process.
2.2.3.6 End-state
Ideally, the end-state would represent complete proficiency and native-like command of the newly acquired language (Klein, 1990:50). However, this is not the case for many learners as fossilization sets in before the learner has mastered the language.
24
Fossilization can be described as a premature termination of the progress of a certain domain–such as pronunciation–while the learner continues to develop in the remaining domains–such as morphology and syntax (Klein, 1990:51; Siegel, 2009:584; Schneider, 2013:144). Fossilization occurs selectively, and varies grossly between individuals. However, one issue with the notion of fossilization is an apparent lack of stability of performance, which is why some researchers in SLA prefer to avoid the term fossilization (Birdsong, 2009:403). Occasionally a learner may also experience backsliding (Klein, 1990:52; Mesthrie & Bhatt, 2008:170), which occurs when a learner suddenly, during an exchange and only for a short duration, regresses into an earlier stage of acquisition and makes errors.
The above dimensions further illustrate that the process of SLA is a complex and challenging undertaking. Factors in the environment, such as the opportunity to communicate with others, but also factors within the individual, such as motivation and attitude, play a role in the structure and speed at which a person may acquire a language. However, this only represents a part of SLA. Even in an ideal environment with access to the target language and accommodating speakers, wherein the learner is fully motivated, invested, and has an open attitude toward the target language; learners are faced with four tasks that they must overcome to reach complete proficiency. 2.2.4 The four problems for learners
Learners of a second language encounter four main tasks in their transition into a fluent speaker. As noted in section 2.2.2, the language of the learner at various levels of competence are referred to as learner varieties of language. The learner transitions from one variety (or system) to one which is more advanced. Thus, the four tasks discussed below can be viewed as transitional steps. Once the learner is able to overcome the first problem, they transition into a new level of competence and can now address the second problem. A short summary of these four problems is provided here; for a full exposition refer to Klein (1990).
